Collect data and load libraries

library("ggplot2")
library("plotly")

keggdf <- read.csv("../kegg_analysis/kegg_to_path_output_null.txt", sep = "\t")

Sample of our data frame

head(keggdf)
##                          KEGG.Pathway Leaf58 pMPPla107
## 1            Global and overview maps     20        24
## 2                     Drug resistance   null         1
## 3                     Immune diseases      1         1
## 4 Signaling molecules and interaction   null         1
## 5                  Cellular community   null         1
## 6               Nucleotide metabolism     12        19

Plotly pie charts

Plotly offers interactive charts when rendered in HTML. These charts show the counts of each KEGG pathway post UProC analysis.

#Note the showlegend option is set to false.  After an update from plotly this needed to be added to avoid the legend crowding the chart for this format.  However, just remove that command if the legend is desired.

pMP_pie <- plot_ly(keggdf, labels = ~KEGG.Pathway, values = ~pMPPla107, type = 'pie',textposition = 'outside',textinfo = 'value') %>%
  layout(title = 'pMPPla107 KEGG Pathways', showlegend = FALSE,
         xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
         yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
pMP_pie
pMPPla107_pie 
L58_pie <- plot_ly(keggdf, labels = ~KEGG.Pathway, values = ~Leaf58, type = 'pie',textposition = 'outside',textinfo = 'value') %>%
  layout(title = 'pBASL58 KEGG Pathways', showlegend = FALSE,
         xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
         yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
L58_pie
pBASL58_pie

Note: to add legend in R add ‘showlegend=TRUE’ after ‘title=“”,’. When rednering in markdown the legend overlaps the figure so I left it out here.

Saving the image

As previously mentioned plotly has some bugs, another of which includes the inability to export images within R studio. Therefore, another package is used for this task.

library("webshot")
export(pMP_pie, file = "imgage_name.png")
export(L58_pie, file = "imgage_name2.png")

Creating a lollipop graph

For those that prefer a different graph to pie charts, lollipop graphs offer a nice visualization.

#Read in a comma serperated file
df <- read.csv("../kegg_analysis/kegg_output_lolipop_rm_some_paths.csv", header =TRUE, sep = ",")

head(df)
##                    Pathway Count   Plasmid
## 1   Replication and repair    26 pMPPla107
## 2 Global and overview maps    24 pMPPla107
## 3   Replication and repair    22   pBASL58
## 4 Global and overview maps    20   pBASL58
## 5    Nucleotide metabolism    19 pMPPla107
## 6    Nucleotide metabolism    12   pBASL58
#Plot a grouped lolipop graph
ggplot(df)+
  geom_linerange(aes(x = reorder(Pathway, Count), ymin = 0, ymax = Count, colour = Plasmid), 
                 position = position_dodge(width = 0.5))+
  geom_point(aes(x = reorder(Pathway, Count), y = Count, colour = Plasmid),
             position = position_dodge(width = 0.5))+
  coord_flip() +
  xlab("KEGG Pathways")+
  ylab("Number of Predicted Genes in Pathway")+
  theme(axis.title = element_text(face = "bold"), legend.position = c(0.85,0.15))

Last updated 12/11/18